JOURNAL OF CELLULAR PHYSIOLOGY 207:767–774 (2006)

Subcutaneous Tissue Fibroblast Cytoskeletal

Remodeling Induced by Acupuncture:
Evidence for a Mechanotransduction-Based Mechanism
HELENE M. LANGEVIN,1* NICOLE A. BOUFFARD,1 GARY J. BADGER,2
DAVID L. CHURCHILL,1AND ALAN K. HOWE3
1
Department of Neurology, 2Department of Medical Biostatistics,
3
Department of Pharmacology, Vermont Cancer Center,
University of Vermont College of Medicine, Burlington Vermont

Acupuncture needle rotation has been previously shown to cause specific mechanical stimulation of subcutaneous connective
tissue. This study uses acupuncture to investigate the role of mechanotransduction-based mechanisms in mechanically-induced
cytoskeletal remodeling. The effect of acupuncture needle rotation was quantified by morphometric analysis of mouse tissue
explants imaged with confocal microscopy. Needle rotation induced extensive fibroblast spreading and lamellipodia formation
within 30 min, measurable as an increased in cell body cross sectional area. The effect of rotation peaked with two needle
revolutions and decreased with further increases in rotation. Significant effects of rotation were present throughout the tissue,
indicating the presence of a response extending laterally over several centimeters. The effect of rotation with two needle revolutions
was prevented by pharmacological inhibitors of actomyosin contractility (blebbistatin), Rho kinase (Y-27632 and H-1152), and Rac
signaling. The active cytoskeletal response of fibroblasts demonstrated in this study constitutes an important step in understanding
cellular mechanotransduction responses to externally applied mechanical stimuli in whole tissue, and supports a previously
proposed model for the mechanism of acupuncture involving connective tissue mechanotransduction. J. Cell. Physiol. 207: 767–
774, 2006. ß 2006 Wiley-Liss, Inc.

Subcutaneous tissue is part of a network of ‘‘loose’’
connective tissue extending throughout the body includ-
ing fasciae and interstitial connective tissue. This web of
tissue is populated by an interconnected network of
fibroblasts that rapidly respond to tissue stretch (within
minutes) with active, dynamic, and reversible changes
in cell shape (Langevin et al., 2004, 2005). Unlike dermis
and load bearing connective tissues (e.g., ligament,
tendon), subcutaneous tissue has a low tensile modulus
close to that of cells (Iatridis et al., 2003). Thus
fibroblasts within loose connective tissues may perceive
a greater range of externally applied forces than those
embedded in a denser collagen matrix. These recent
findings suggest that loose connective tissue may
actively and rapidly respond to changing tissue loads.
Understanding the cellular mechanisms underlying
these responses is central to establishing the nature
of this new and potentially important physiological
function.
The cytoskeleton has emerged as a key structural
element allowing transmission of externally applied
mechanical forces to the cell and conversion of these
forces into biochemical responses (Chicurel et al., 1998,
2003). Indeed, the cytoskeleton’s plasticity, mechanical
responsiveness, and links to key intracellular regula-
tory proteins such as Rho GTPases form the basis of
mechanotransduction (Banes et al., 1995; Hall, 1998).
Because mechanotransduction concerns interactions
of the cell with its environment, studying mechano-
transduction mechanisms in whole tissues is an impor-
tant complement to cell culture models. This is
especially important for connective tissue, since its
extracellular matrix and mechanical environment
are much more complex than even ‘‘3-d’’ cell culture
ß 2006 WILEY-LISS, INC.
conditions (Grinnell, 2003). In this context, the ancient
technique of acupuncture provides a useful experimen-
tal tool to investigate mechanotransduction within
connective tissue. It was recently shown that, during
acupuncture, rotation of the acupuncture needle causes
winding of subcutaneous tissue (but not dermis), and
pulling of the loose collagen bundles from the periphery
toward the needle (Langevin et al., 2001a). In rats,
needle rotation produced measurable changes in sub-
cutaneous tissue matrix architecture together with a
tenfold increase in the amount of force necessary to pull
the needle out of the tissue (Langevin et al., 2002). A
unique feature of acupuncture therefore is that, unlike
stretching of whole skin, needle rotation specifically
probes the loose subcutaneous tissue layer. Thus,
acupuncture allows investigation of cellular responses
to a highly-specific mechanical stimulus.
In this study, we have used acupuncture to investi-
gate the role of mechanotransduction-based signaling
mechanisms in mechanically-induced cytoskeletal
Contract grant sponsor: National Center for Complementary and
Alternative Medicine (NCCAM); Contract grant number: RO1
AT01121.
*Correspondence to: Dr. Helene M. Langevin, Department of
Neurology, University of Vermont College of Medicine, Given
C423, 89 Beaumont Avenue, Burlington, VT 05405.
E-mail: helene.langevin@uvm.edu
Received 7 November 2005; Accepted 10 January 2006
DOI: 10.1002/jcp.20623
768 LANGEVIN ET AL.

remodeling. We hypothesized that, in mouse whole skin

tissue explants, acupuncture needle rotation induces
dose-dependent morphological changes in subcutaneous
tissue fibroblasts, and that these changes involve Rho
and Rac signaling mechanisms as well as actomyosin
contractility. In addition to investigating basic cellular
physiological processes, this approach may eventually
contribute to a better understanding of how connective
tissue mechanotransduction may be part of acupunc-
ture’s therapeutic mechanism.

MATERIALS AND METHODS

Experiment setup and protocol
Experimental protocols were approved by the University of
Vermont IACUC. C57Black-6 mice (19–24 g) were sacrificed by
decapitation. Immediately after death, an 8 cm by 3 cm tissue
flap containing dermis, subcutaneous muscle, and subcuta-
neous tissue was excised from the back of each mouse. Excision
involved minimal cutting of the loose cleavage plane separat-
ing the subcutaneous tissue layer from the back muscles.
Tissue flaps were placed transversely in grips and immersed in
an incubation bath containing 378C HEPES-physiological
saline pH 7.4 as previously described and placed vertically in
a holder with the proximal grip connected to a 500g (4.9 N)
capacity load cell (Langevin et al., 2005). The tissue was
elongated at a rate of 1 mm/sec by advancing a micrometer
connected to the distal tissue grip until a varying amount of
preload was achieved. The tissue length was then fixed by
attaching a pair of stabilization bars. Tissue and grips together
were then disconnected from the load cell and placed under a
dissecting microscope. An acupuncture needle (Seirin, Japan,
0.25 mm diameter, and 50 mm length) was inserted parallel to
the dermis, carefully ‘‘tunneling’’ the needle through the
middle of the subcutaneous tissue layer (Fig. 1A). Extreme
care was taken during this procedure to insert the needle
without pulling on the tissue. The tissue and grips were then
placed back into the incubation bath. A modified version of a
computer-controlled acupuncture needling instrument (Lan-
gevin et al., 2001b) was used to perform needle rotation (Fig. 1).
The instrument’s motor was attached to the needle and placed
into a clamp to maintain the needle in a fixed vertical position
during rotation (Fig. 1B). The needle was rotated immediately
(to mimic the clinical situation in which needles are usually
manipulated directly following insertion) with a varying
number of clockwise needle revolutions (or not rotated) at a
constant speed (1 rev/sec) and acceleration of 3 revs/sec2. After
needle rotation, the motor was disconnected and the tissue was
further incubated for 30 min. The tissue was then immersion
fixed in 95% ethanol for 1 h, then rinsed overnight in Fig. 1. Experimental setup. A: Excised mouse tissue after preload
phosphate-buffered saline (PBS) /1.0% Bovine serum albumin and during the acupuncture needle insertion procedure. B: Excised
at 48C with the needle in place. mouse tissue is maintained at 378C at a fixed length after setting
tissue preload. An acupuncture needle inserted into the subcutaneous
tissue is attached to a needle grip and motor (needle grips placed
Experimental design and variables vertically in bath). C: Location of subcutaneous tissue samples
relative to the needle (needle parallel to skin).
Varying preload with no rotation. To establish the
optimal tissue preload for needle rotation experiments (defined
as the maximum amount of preload that would not by itself addition of inhibitors was followed by a 10 min rest, then
cause detectable cellular changes), tissue flaps were elongated preload and needle rotation as above.
until a load of, 1.96 mN (equivalent to a 0.2 g mass), 2.9 mN
(0.3 g), or 4.9 mN (0.5 g) was achieved.
Histochemical staining, confocal microscopy,
Constant preload with varying rotation. Tissue flaps
were elongated to a preload of 2.9 mN followed by either, 0, 2, 4, and morphometric analysis
8, or 12 clockwise needle revolutions. Following the overnight rinse (described above), the needle
Pharmacological experiments with two needle revolu- was removed from the tissue and three subcutaneous tissue
tions. Rho signaling and associated actomyosin contractility samples (each 10
10 mm) were dissected from the tissue flap.
was investigated with blebbistatin (100 mM) (Calbiochem, San Each sample was placed flat on a glass slide. The sample that
Diego, CA) (Straight et al., 2003) and two different inhibitors of had included the needle was labeled ‘‘medial,’’ and the other
Rho kinase: Y-27632 (10 mM) (Biomol, Plymouth Meeting, PA), samples labeled ‘‘lateral right’’ and ‘‘lateral left,’’ respectively
and H-1152 (10 mM) (Calbiochem). Rac signaling was inhibited (Fig. 1C). The slides were stained with Texas Red conjugated
with NCS23766 (117 mM) (Calbiochem). As a control, SP600125 phalloidin (a specific stain for polymerized actin) and SYTOX
(10 mM) (Calbiochem) was used a JNK-2 signaling inhibitor not nuclear stain, then imaged with a Bio-Rad MRC 1024 confocal
involved in cytoskeletal remodeling. All inhibitors were microscope (Bio-Rad Microsciences, Hercules, CA), as pre-
solubilized directly in 378 HEPES PSS (similar results for viously described (Langevin et al., 2004). Each sample was
SP600125 were obtained with and without DMSO). The divided longitudinally into three roughly equal-sized areas.
Journal of Cellular Physiology DOI 10.1002/jcp
 FIBROBLAST RESPONSE TO ACUPUNCTURE
One field was imaged in the center of each area (nine fields per
animal). Fields were chosen at low power by an individual blind
to the study variable (preload or number of revolutions).
Imaged fields were located in the center of each area without
regard for needle position, since the needle track was not
always visible. For each field, a stack of 20 (313
313 mm)
images was acquired at a 1 mm inter-image interval. Image
stacks were imported into the analysis software package
MetaMorph (version 6.0; Universal Imaging Corporation,
Downington, PA) for morphometric analysis. Cell body cross
sectional area was measured as previously described (Lange-
vin et al., 2005).
Statistical methods
Two-way analyses of variance (ANOVA) were used to test for
differences in mean cell body cross sectional area in each
experiment. In the first experiment, designed to examine the
effect of varying preload, the two factors were preload [an
across-subject factor with three levels [(0.2, 0.3, and 0.5 g)] and
region [a within-subject factor with two levels (medial vs.
lateral)]. For the second experiment designed to examine the
effects of varying the number of needle revolutions, the two
factors were number of revolutions [an across-subject factor
with five levels [(0, 2, 4, 8, and 12)] and region [a within-subject
factor with two levels (medial vs. lateral)]. Additional ANO-
VA’s were performed to compare experimental conditions done
at two needle revolutions in the presence of pharmacological
inhibitors to those without inhibitors. In all experiments, data
from lateral right and lateral left samples were combined into
one lateral region. When the F-test from the ANOVA was
significant (P < 0.05), Fisher’s LSD was used to perform
comparisons among means. Statistical analyses were per-
formed using SAS statistical Software Version 8.02.
RESULTS
Effect of varying preload
Preload had a significant effect on cell body cross
sectional area (F2,8 ¼ 12.3, P ¼ 0.004). Cross sectional
area did not significantly change until a preload of
4.9 mN was applied to the tissue (Fig. 2). There was
no evidence of significant differences in cell body cross
sectional area between the regions of the tissue
(F1,8 ¼ 3.4, P ¼ 0.10), nor was there evidence that the
effect of preload was different across regions (F2,8 ¼ 0.4,
P ¼ 0.69). Based on these results, we chose 2.9 mN as our
preload in subsequent experiments investigating the
effects of varying rotation.
Fig. 2. Mean cell body cross sectional area with varying tissue
preload. Values sharing a common letter are not significantly
different. Error bars represent SE.
Journal of Cellular Physiology DOI 10.1002/jcp
769
Effect of needle rotation
Two needle revolutions caused fibroblast cell bodies to
become large and ‘‘sheet-like’’ with extensive spreading
and lamellipodia formation (Fig. 3B,D). This was in
contrast to the small cell bodies and long cytoplasmic
processes seen with no rotation (Fig. 3A,C).
Statistical analysis of morphometric measurements
showed that the number of needle revolutions had a
significant effect on fibroblast cell body cross sectional
area (F4,28 ¼ 6.9, P < 0.001) (Fig. 4A, Table 1). Cell body
cross sectional area peaked with two needle revolutions
and decreased with further increases in rotation. The
increase in mean cross sectional area from 0 to 2 needle
revolutions was similar in magnitude to the difference
produced by increasing the preload from 2.9 to 4.9 mN
(Figs. 2 and 4).
With no rotation, fibroblast morphology was uniform
throughout the tissue (Fig. 5A2 –E2). With rotation, the
largest cells were seen in the medial samples
(Fig. 5B1,D1), with the exception of a 1 or 2 mm-wide
band immediately adjacent to the needle where a
‘‘whorl’’ of connective tissue was often seen containing
thin, elongated, spindle-shaped fibroblasts that were
frequently distorted with processes that appeared
broken and/or recoiled (Fig. 5, compare Fig 5C1,C2).
Morphometric examination of separate tissue regions
revealed that the effect of rotation was somewhat more
pronounced in the tissue samples closest to the needle
(Fig. 4B, open bars), though there was no evidence
that the rotation effect was region-specific (F4,18 ¼ 2.2,
P ¼ 0.10).
Effect of pharmacological inhibitors
Based on the above results, pharmacological experi-
ments were performed with two needle revolutions.
Fibroblasts incubated with rotation in the presence of
blebbistatin, Rho kinase inhibitors (Y-27632 and H-
1152), and the Rac 1 inhibitor had small cell bodies
(Fig. 6B–E). In contrast, fibroblasts incubated with
rotation in the presence of the JNK-2 inhibitor had large
cell bodies, similar in appearance to those incubated in
the absence of inhibitors with needle rotation (Fig. 6F).
Morphometric measurements showed that, when needle
rotation was performed in the presence of blebbistatin,
Rho kinase or Rac-1 inhibitors, mean cell body cross
sectional area was significantly smaller compared
with needle rotation without inhibitors (Fisher’s LSD)
(Fig. 7, Table 1). Thus, the increase in fibroblast cross-
sectional area induced by acupuncture needle rotation
was prevented by inhibition of actomyosin contractility
as well as Rho and Rac signaling, but not JNK-2
signaling.
DISCUSSION
Acupuncture needle rotation caused fibroblast
spreading and lamellipodia formation involving Rho
and Rac signaling as well as actomyosin interaction. The
results of this study therefore show that mechanical
stimulation of subcutaneous tissue causes active fibro-
blast cytoskeletal remodeling involving mechanotrans-
duction-based mechanisms. It is now well established
that the cytoskeleton is an integrated and dynamic
system within the cell that actively interacts with the
extracellular matrix via specialized sites on the cell
surface (actin-integrin focal adhesions) (Lauffenburger
and Horwitz, 1996; Geiger and Bershadsky, 2001).
Changes in cell shape (e.g., cell spreading during
migration, neuronal growth cone extension, or cellular
770 LANGEVIN ET AL.

Fig. 3. A, B: Mouse subcutaneous tissue incubated for 30 min after acupuncture needle rotation (two
revolutions) (B) compared with no rotation (A). C, D: Individual fibroblasts with rotation (D) and without
rotation (C). All samples were stained with Texas-red conjugated phalloidin counter stained with SYTOX
nucleic acid stain, and imaged with confocal microscopy. Scale bars, 40 mm. A and B are composite
projections of stacks containing 20 optical sections taken at 1 mm intervals. C and D are projections of
relevant optical sections containing the cells. [Color figure can be viewed in the online issue, which is
available at www.interscience.wiley.com.]

response to mechanical substrate deformation) are
accompanied by active and coordinated reorganizations
of the cytoskeleton. These events include actin filament
polymerization/depolymerization, microtubule assem-
bly/disassembly, actomyosin contraction, and focal
adhesion formation/disruption (Theriot and Mitchison,
1991; Ponti et al., 2004). Several Rho GTPase signaling
molecules are known to play key roles in coordinating
these processes (Rottner et al., 1999; Ridley, 2001). Rac
induces the formation of focal adhesion complexes along
the edge of lamellipodia, while Rho activates the
transformation of these focal complexes into focal
contacts via generation of myosin-II dependent tension
(Geiger and Bershadsky, 2001). Recent evidence in
addition suggests that Rho is involved in microtubule
targeting, capture, and stabilization at focal adhesion
sites (Ishizaki et al., 2001; Wittmann and Waterman-
Storer, 2001) and that microtubules migrate toward
focal adhesions in areas of externally applied tensile
stress (Kaverina et al., 2002). Dynamic cytoskeletal
mechanisms also are thought to play an important role
in cell biomechanics by balancing internal tensile and
compressive cellular forces with forces externally
applied to the cell via the extracellular matrix (Wang
et al., 2002; Ingber, 2003). According to this ‘‘cellular
tensegrity’’ model, actomyosin contraction coupled to
microfilaments generate cellular internal tension,
while the microtubule network provides internal com-
pressive forces (although some controversy exists as to
whether microtubules are rigid-enough to fulfill this
Journal of Cellular Physiology DOI 10.1002/jcp
function) (Mitchison and Kirschner, 1984; Gittes et al.,
1993; Mickey and Howard, 1995; Waterman-Storer
and Salmon, 1997; Gupton et al., 2002; Stamenovic
et al., 2002).
The active cellular responses found in this study are
consistent with the above biochemical and biomechani-
cal models of cytoskeletal function, and suggest the
following sequence of events in response to acupuncture
needle rotation: (1) winding and pulling of tissue from
the periphery toward the needle; (2) initial pull of
extracellular matrix on fibroblasts at existing focal
contacts; (3) formation of lamellipodia (Rac-induced) in
regions of the cell that are mechanically stimulated
(predominantly in the plane of the pulled tissue); (4)
increased actomyosin contraction (Rho-induced) with-
out distinct stress fiber formation (due to the complex
three-dimensional pattern of matrix attachments); (5)
microtubule migration and stabilization; (6) increased
intracellular tension, cell expansion, and flattening in
the tissue plane until a new tension equilibrium is achi-
eved between intracellular tension (actomyosin-driven)
and two types of opposing forces: (a) extracellular matrix
counter-tensional forces and (b) intracellular compres-
sive forces provided by the expanded cytoskeleton. This
proposed mechanism is illustrated in Figure 8. Such a
mechanism, situated at the nexus of current biochem-
ical and biomechanical understanding of mechanotrans-
duction, would not only explain the cell shape changes
observed in this study, but would also support the
previously proposed concept that connective tissue
 FIBROBLAST RESPONSE TO ACUPUNCTURE 771

with its known functions (activation gene expression

Fig. 4. Effect of increasing amounts of acupuncture needle rotation
on mean fibroblast cell body cross sectional area. A: Overall effect of
rotation in combined medial and lateral tissue samples. Values
sharing a common letter are not statistically significant. B: Cell body
cross-sectional area in medial (open bars) and lateral (solid bars)
tissue samples. Error bars represent SE.
tension is actively regulated by fibroblasts (Eastwood
et al., 1996).
An important effect of mechanotransduction is the
regulation of mechanosensitive genes via signaling
pathways linked to mechanically-induced cytoskeletal
reorganization (Chiquet et al., 2003). Fibroblast cytos-
keletal remodeling in response to acupuncture therefore
may have effects on mechanotransduction-mediated
gene expression and connective tissue matrix remodel-
ing. Change in extracellular matrix composition days to
weeks following acupuncture may cause modulation of
sensory input from mechanosensory and nociceptor
afferent neurons within connective tissue. As expected,
acupuncture-induced changes in cell shape were not
prevented by inhibition of JNK-2, which is consistent
downstream of Rac and Cdc 42, and regulation of
apoptosis (Coso et al., 1995; Minden et al., 1995)).
Whether this type of mechanical stimulus causes JNK-
2-mediated changes in gene expression will be an
important topic for future studies.
We believe that an important aspect of this study
is that it can serve to bridge the extensive knowledge
on mechanotransduction derived from cell culture
models to an eventual understanding of (1) the effect of
mechanical forces on cells in vivo in both normal
and diseased connective tissue, (2) how mechanical
forces of a certain type, amplitude, and duration can be
therapeutic, and (3) what ‘‘dose’’ of mechanical stimula-
tion produces the optimal therapeutic effect. The
fibroblast spreading induced by needle rotation was
similar to that previously reported with stretching of
subcutaneous tissue together with dermis (Langevin
et al., 2005), showing that two different types of
mechanical stimuli (tissue stretch and acupuncture)
caused similar subcutaneous tissue fibroblast morpho-
logical responses. Thus, connective tissue mechano-
transduction responses may be common to different
types of therapeutic interventions using externally
applied mechanical forces (e.g., physical therapy, mas-
sage, and acupuncture).
The importance of acupuncture needle manipulation
techniques has been stressed in acupuncture texts
since ancient times (Yang, 1991; Wu, 1993). These
techniques typically consist of varying combinations of
rotational and axial needle movements. The frequency,
amplitude, and duration of these needle movements
traditionally are believed key to optimize treatment
outcome (O’Connor et al., 1981; Cheng, 1987). In this
series of experiments, two revolutions of the acupunc-
ture needle caused the greatest increase in fibroblast cell
body cross sectional area. This amount of needle rotation
caused cellular changes comparable to those of a
surprisingly small amount of tissue load (1.96 mN
(0.2 g)). This pattern of cellular response is reminiscent
of biomechanical effects previously described in bone
where smaller strain levels caused greater biological
effects than larger ones (Brand and Stanford, 1994).
Further studies will be important to examine cellular
responses to smaller amounts of rotation between zero
and two revolutions. The decreased spreading of
fibroblasts with greater than two revolutions, on the
other hand, may be due to cells sensing ‘‘too much’’ (or
too rapid) external tension and letting go of existing
adhesions, thus decreasing or delaying Rac-induced
lamellipodia formation. Such a pattern (initial retrac-
tion followed cell respreading) was described in cultured
cells in response to matrix traction (Petroll et al., 2004).
It is also possible that the decreased cell spreading with
increasing rotation was due to increased slippage at the

TABLE 1. Fibroblast corss sectional area measurements

#Needle revolutions
Pharmacological
inhibition Drug 0 2 4 8 12

None None 348.4  92.4 (8) 611.2  141.0 (7) 510.3  90.3 (6) 405.9  93.1 (8) 435.5  106.2 (4)
Actomyosin Blebbistatin — 392.1  44.4 (3) — — —
Rac 1 NSC23766 — 316.8  118.9 (3) — — —
Rho kinase Y27632 — 346.8  25.4 (3) — — —
Rho kinase H-1152 — 323.8  86.2 (3) — — —
JNK 2 Calbiochem 420119 — 492.9  76.7 (3) — — —

Results presented as mean  SD fibroblast cross-sectional area in combined (media and lateral) tissue samples. Value in parentheses (Nanimal); number of cells
measured per animal ranged from 43 to 60 cells. Measurements are expressed in mm2.

Journal of Cellular Physiology DOI 10.1002/jcp

772 LANGEVIN ET AL.

Fig. 5. Fibroblast morphology in different parts of the tissue with
and without needle rotation (two revolutions). Tissue sample locations
(A–E) are shown in top panel. With rotation, fibroblasts in tissue
nearest the needle were spindle-shaped with processes that appeared
undulating, twisted, or broken (C1), while fibroblasts in the rest of the
tissue were large and ‘‘sheet-like’’ (A1, B1, C1, D1). The largest cells
with rotation were found within a zone 1–5 mm away from the needle
(B1,D1). With no rotation, fibroblasts throughout the tissue had a
similar ‘‘dendritic’’ appearance with small cell bodies and long
branching processes (A2 –E2). Scale bars, 40 mm. [Color figure can be
viewed in the online issue, which is available at www.interscience.
wiley.com.]

needle/tissue interface causing decreased tissue wind- needle torque, we did collect indirect torque measure-
ing. We consider this unlikely based on the torque ments by recording the amount of current passing
developing at the needle/tissue interface during rota- through the motor during rotation. Mean (SD) needle
tion. Although the aim of this study was not to measure torque during four and eight needle revolutions were

Fig. 6. Fibroblast morphology with needle rotation with and without pharmacological inhibitors. All
samples received two needle revolutions. A: Control without inhibitor; (B) actomyosin contractility
inhibitor (blebbistatin); (C) Rac-1 inhibitor (NCS23766); (D) Rho kinase inhibitor (Y27632), (E) Rho
kinase inhibitor (H-1152); (F) JNK-2 inhibitor (SP600125). [Color figure can be viewed in the online issue,
which is available at www.interscience.wiley.com.]

Journal of Cellular Physiology DOI 10.1002/jcp

 FIBROBLAST RESPONSE TO ACUPUNCTURE
Fig. 7. Effect of pharmacological inhibitors on mean fibroblast cross
sectional area, compared with 0 and 2 needle revolutions without
inhibitors. All samples with inhibitors received two needle revolu-
tions: Blebbistatin (actomyosin contractility inhibitor); Rac-1 inhibitor
(NCS23766); Rho kinase inhibitor (Y27632), Rho kinase inhibitor (H-
1152); JNK-2 inhibitor (SP600125). *indicates significant difference
from two needle revolutions without inhibitor; {indicates significant
difference from 0 needle revolutions without inhibitor. Error bars
represent SE.
0.21  0.14 and 0.77  0.05 mNm, respectively. Torque
during 2 needle revolutions was too low to be reliably
detectable and 12 revolutions produced torques that
were too large and saturated the motor current. This
progressive increase in torque from 0 to 12 rotations
therefore suggests that, in these experiments, the
smaller cell cross sectional areas with increasing rota-
tion was not due to the cells perceiving less mechanical
force due to slippage. Although additional studies will be
needed to determine whether a causal link exists
between fibroblasts’ response to mechanical stimulation
and therapeutic effects (e.g., pain reduction), the dose-
related physiological effects observed in this study may
eventually prove to be important clinically.
Significant effects of rotation were present through-
out the tissue, indicating the induction of an active
response extending laterally over several centimeters.
The largest cells were measured in the medial tissue
samples corresponding to a 0.5 cm zone on both sides
of the needle. Since the fibroblasts measured in the
medial samples also included some of the thin spindle-
shaped fibroblasts found in the immediate vicinity of
the connective tissue ‘‘whorl’’ (Fig. 4C1), it is likely that
Fig. 8. Summary illustration representing proposed active fibroblast
response to acupuncture needle rotation. Needle rotation (B) causes
winding of collagen fibers around the needle and formation of a
‘‘whorl’’ of collagen and fibroblasts in the area immediately surround-
ing the needle. With small amounts of needle rotation, pulling of
collagen fibers towards the needle causes fibroblasts further away
from the needle to respond by changing shape, becoming large and
‘‘sheet-like’’ in marked contrasts with the small cell bodies and long
Journal of Cellular Physiology DOI 10.1002/jcp
773
mean cell cross sectional area in the medial tissue
samples would have been even greater if these cells had
been excluded. In order to mimic the clinical situation,
the needle was not held in the final rotation position
after the completion of rotation, but rather was allowed
to remain free in the tissue during the dwell time. We did
observe some amount of ‘‘spinning back’’ of the needle
immediately after disconnection, but did not assess this
quantitatively. The remaining whorl of connective
tissue grossly observable after fixation suggests that
the needle does not spin back completely, but that some
amount of tissue winding remains present throughout
the 30 min dwell time between needle rotation and
tissue fixation.
A limitation of this study is that, since our imaging
methods required tissue fixation, we could only assess
cellular effects at one time point following needle
rotation. We chose a 30 min dwell time between needle
rotation and tissue fixation based on (1) a previous study
of tissue stretch (Langevin et al., 2005) in which
measurable effects of stretch occurred between 10 and
120 min after stretching the tissue and (2) because
acupuncturists typically leave needles in place for 20–
30 min after manipulation. Other possible effects of
needle rotation that may occur between 0 and 30 min
while the needle is still in place (e.g., increase in
intracellular calcium or release of ATP) could be
examined in future studies using live cell imaging
techniques. Such effects may, for example, result in
cell-to-cell signaling, which may contribute to spreading
of the effect of acupuncture along connective tissue
planes.
In acupuncture practice, the angle of needle insertion
relative to the skin surface varies widely from very
shallow (0–108) to perpendicular (908) depending on the
clinical situation (Cheng, 1987). Because mouse sub-
cutaneous tissue is very thin (50–100 mm), we chose to
perform our quantitative experiments using a needle
angle of 08 (needle parallel to the skin surface) in order
to maximize the precision and consistency of needle
placement within subcutaneous tissue and thus
the reproducibility of our testing conditions. However,
we obtained similar results using both parallel and
perpendicular needle orientations (data not shown)
suggesting that the basic phenomena observed in this
study are present using a wide range of needling angles.
In summary acupuncture induced an active cytoske-
letal response in connective tissue fibroblasts with
extensive cell spreading and lamellipodia formation
branching processes (‘‘dendritic’’ morphology) seen without needle
rotation (A). After needle rotation, a new tension equilibrium is
achieved between actomyosin-driven intracellular tension (intracel-
lular white arrows) and two types of opposing forces: extracellular
matrix counter-tensional forces (extracellular black arrows) and
intracellular compressive forces provided by the expanded cytoskele-
ton (intracellular black arrows). Gray dots represent focal contacts.
774 LANGEVIN ET AL.
involving Rho kinase, Rac signaling, and actomyosin
contraction. This effect was measurable as an increase
in fibroblast cross sectional area, which was greatest
with two needle revolutions and decreased with further
increases in rotation. A significant effect of rotation was
present several centimeters away from the needle. This
study constitutes an important step in understanding
cellular mechanotransduction responses to externally
applied mechanical forces in whole connective tissue.
Further studies will address downstream effects of this
cellular response (such as gene expression) leading to
long term tissue changes that may translate into
therapeutic mechanisms relevant to a wide range of
treatments including physical therapy and massage as
well as acupuncture.
ACKNOWLEDGMENTS
The authors thank Kirsten N. Storch and Alexander
T. Danco for technical assistance, Debbie Stevens-Tuttle
for help in preparing the manuscript, and Dr. Kenneth
R. Cutroneo and Dr. Marilyn J. Cipolla for helpful
discussions. This study was funded by the National
Center for Complementary and Alternative Medicine
Research Grant RO1-AT01121. Its contents are solely
the responsibility of the authors and do not necessarily
represent the official views of the National Center for
Complementary and Alternative Medicine, National
Institutes of Health.
LITERATURE CITED
Banes AJ, Tsuzaki M, Yamamoto J, Fischer T, Brigman B, Brown T, Miller L.
1995. Mechanoreception at the cellular level: The detection, interpretation, and
diversity of responses to mechanical signals. Biochem Cell Biol 73(7–8):349–
365.
Brand RA, Stanford CM. 1994. How connective tissues temporally process
mechanical stimuli. Med Hypotheses 42(2):99–104.
Cheng X. 1987. Chinese acupuncture and moxibustion. In: Cheng X, editor.
Chinese acupuncture and moxibustion, 1st edition. Beijing: Foreign Language
Press. pp 322–329.
Chicurel ME, Chen CS, Ingber DE. 1998. Cellular control lies in the balance of
forces. Curr Opin Cell Biol 10(2):232–239.
Chiquet M, Renedo AS, Huber F, Fluck M. 2003. How do fibroblasts translate
mechanical signals into changes in extracellular matrix production? Matrix
Biol 22(1):73–80.
Coso OA, Chiariello M, Yu JC, Teramoto H, Crespo P, Xu N, Miki T, Gutkind JS.
1995. The small GTP-binding proteins Rac1 and Cdc42 regulate the activity of
the JNK/SAPK signaling pathway. Cell 81(7):1137–1146.
Eastwood M, Porter R, Khan U, McGrouther G, Brown R. 1996. Quantitative
analysis of collagen gel contractile forces generated by dermal fibroblasts and
the relationship to cell morphology. J Cell Physiol 166(1):33–42.
Geiger B, Bershadsky A. 2001. Assembly and mechanosensory function of focal
contacts. Curr Opin Cell Biol 13(5):584–592.
Gittes F, Mickey B, Nettleton J, Howard J. 1993. Flexural rigidity of microtubules
and actin filaments measured from thermal fluctuations in shape. J Cell Biol
120(4):923–934.
Grinnell F. 2003. Fibroblast biology in three-dimensional collagen matrices.
Trends Cell Biol 13(5):264–269.
Gupton SL, Salmon WC, Waterman-Storer CM. 2002. Converging popu-
lations of f-actin promote breakage of associated microtubules to spatially
Journal of Cellular Physiology DOI 10.1002/jcp
regulate microtubule turnover in migrating cells. Curr Biol 12(22):1891–
1899.
Hall A. 1998. Rho GTPases and the actin cytoskeleton. Science 279(5350):509–
514.
Iatridis JC, Wu J, Yandow JA, Langevin HM. 2003. Subcutaneous tissue
mechanical behavior is linear and viscoelastic under uniaxial tension. Connect
Tissue Res 44(5):208–217.
Ingber DE. 2003. Tensegrity I. Cell structure and hierarchical systems biology.
J Cell Sci 116(Pt 7):1157–1173.
Ishizaki T, Morishima Y, Okamoto M, Furuyashiki T, Kato T, Narumiya S. 2001.
Coordination of microtubules and the actin cytoskeleton by the Rho effector
mDia1. Nat Cell Biol 3(1):8–14.
Kaverina I, Krylyshkina O, Beningo K, Anderson K, Wang YL, Small JV. 2002.
Tensile stress stimulates microtubule outgrowth in living cells. J Cell Sci
115(Pt 11):2283–2291.
Langevin HM, Churchill DL, Cipolla MJ. 2001a. Mechanical signaling through
connective tissue: A mechanism for the therapeutic effect of acupuncture.
FASEB J 15(12):2275–2282.
Langevin HM, Churchill DL, Fox JR, Badger GJ, Garra BS, Krag MH. 2001b.
Biomechanical response to acupuncture needling in humans. J Appl Physiol
91(6):2471–2478.
Langevin HM, Churchill DL, Wu J, Badger GJ, Yandow JA, Fox JR, Krag MH.
2002. Evidence of connective tissue involvement in acupuncture. FASEB J
16(8):872–874.
Langevin HM, Cornbrooks CJ, Taatjes DJ. 2004. Fibroblasts form a body-wide
cellular network. Histochem Cell Biol 122(1):7–15.
Langevin HM, Bouffard NA, Badger GJ, Iatridis JC, Howe AK. 2005. Dynamic
fibroblast cytoskeletal response to subcutaneous tissue stretch ex vivo and in
vivo. Am J Physiol Cell Physiol 288(3):C747–C756.
Lauffenburger DA, Horwitz AF. 1996. Cell migration: A physically integrated
molecular process. Cell 84(3):359–369.
Mickey B, Howard J. 1995. Rigidity of microtubules is increased by stabilizing
agents. J Cell Biol 130(4):909–917.
Minden A, Lin A, Claret FX, Abo A, Karin M. 1995. Selective activation of the
JNK signaling cascade and c-Jun transcriptional activity by the small GTPases
Rac and Cdc42Hs. Cell 81(7):1147–1157.
Mitchison T, Kirschner M. 1984. Microtubule assembly nucleated by isolated
centrosomes. Nature 312(5991):232–237.
O’Connor J, Bensky D, yi xue yuan SZ. 1981. Acupuncture: A comprehensive text.
Chicago: Eastland Press.
Petroll WM, Vishwanath M, Ma L. 2004. Corneal fibroblasts respond rapidly to
changes in local mechanical stress. Invest Ophthalmol Vis Sci 45(10):3466–
3474.
Ponti A, Machacek M, Gupton SL, Waterman-Storer CM, Danuser G. 2004. Two
distinct actin networks drive the protrusion of migrating cells. Science
305(5691):1782–1786.
Ridley AJ. 2001. Rho family proteins: Coordinating cell responses. Trends Cell
Biol 11(12):471–477.
Rottner K, Hall A, Small JV. 1999. Interplay between Rac and Rho in the control
of substrate contact dynamics. Curr Biol 9(12):640–648.
Stamenovic D, Mijailovich SM, Tolic-Norrelykke IM, Chen J, Wang N. 2002. Cell
prestress II. Contribution of microtubules. Am J Physiol Cell Physiol 282(3):
C617–C624.
Straight AF, Cheung A, Limouze J, Chen I, Westwood NJ, Sellers JR, Mitchison
TJ. 2003. Dissecting temporal and spatial control of cytokinesis with a myosin
II Inhibitor. Science 299(5613):1743–1747.
Theriot JA, Mitchison TJ. 1991. Actin microfilament dynamics in locomoting
cells. Nature 352(6331):126–131.
Wang N, Tolic-Norrelykke IM, Chen J, Mijailovich SM, Butler JP, Fredberg JJ,
Stamenovic D. 2002. Cell prestress I. Stiffness and prestress are closely
associated in adherent contractile cells. Am J Physiol Cell Physiol 282(3):
C606–C616.
Waterman-Storer CM, Salmon ED. 1997. Actomyosin-based retrograde flow of
microtubules in the lamella of migrating epithelial cells influences microtubule
dynamic instability and turnover and is associated with microtubule breakage
and treadmilling. J Cell Biol 139(2):417–434.
Wittmann T, Waterman-Storer CM. 2001. Cell motility: Can Rho GTPases and
microtubules point the way? J Cell Sci 114(Pt 21):3795–3803.
Wu J-N. 1993. Ling shu, or, The spiritual pivot. Washington, D.C. Honolulu,
Hawaii: Taoist Center; Distributed by University of Hawaii Press.
Yang J. 1991. The golden needle: And other odes of traditional acupuncture: Book
two of Yang Jizhou’s grand compendium (1601). Edinburgh, New York:
Churchill Livingstone.